Allianz Partners

Allianz Partners handles over 90 million cases annually across health, automotive, and travel insurance in 30+ countries, but rapid growth left operations fragmented across dozens of disconnected legacy systems with no standardization across markets. Claims handling averaged 29 days, and fewer than 10% of claims resolved within 12 hours — a direct drag on customer retention and margin. The operating model scaled linearly with volume: more cases required proportionally more headcount. Previous RPA initiatives failed to handle the variability, cross-market compliance demands, and end-to-end decision complexity required from first notice of loss through payment. For CEO Tomas Kunzmann, the stakes were existential: intermediaries moving faster threatened to erode half the company's value chain.

Allianz Partners deployed autonomous AI agents from Otera (formerly DeepOpinion) across the full claims lifecycle — from reviewing incoming claims and validating coverage against policy terms, through liability assessment, case settlement, customer response generation, and payment triggering. Rather than running a proof-of-concept, the company went straight to production deployment across UK and DACH markets in 2.5 months, leveraging Otera's reusable agentic architecture components and deep insurance domain expertise. The same agentic framework was extended to invoice processing, handling millions of invoices from 16,500 network partners. Complex or exceptional cases escalate to human specialists via a central control hub, with full audit trails and override capabilities maintained throughout to satisfy regulatory requirements across multiple jurisdictions.

End-to-end automation rates jumped from under 10% to over 90%, and average claims handling time fell from 29 days to 3.5 days — an 88% reduction. Touchless cases now resolve in minutes.

Key metrics:

• Claims processed under 12 hours: <10% → >70%
• End-to-end automation rate: <10% → >90%
• Voice-of-customer score: 4.30 → 4.64
• Target incremental annual profit by 2027: €300M
• Revenue growth target: €10B → €13B

The program also demonstrated that the same agentic architecture could absorb adjacent high-volume workflows — invoice processing across 16,500 partners — without proportional headcount growth, validating the platform as a scalable operational foundation.

• Skip the POC cycle when domain expertise is established — Allianz Partners went live in 2.5 months by treating autonomous AI as a platform investment from day one, not an experiment.
• Design for reuse across markets — standardized agentic architecture components made each subsequent rollout faster and lower-risk than the last.
• Invest in regulatory compliance upfront — engaging third-party legal experts specializing in AI regulation early was critical to deploying in heavily regulated markets at speed.
• Target the full process, not isolated steps — RPA failed because it addressed repetitive tasks without touching the core cycle time; only end-to-end automation moved the headline metric.
• Tie AI outcomes to P&L metrics — framing responsiveness as the driver of both customer retention and cost-per-claim gave the program executive sponsorship and clear accountability.